Experimental analysis of binary differential evolution in dynamic environments

  • Authors:
  • Alp Emre Kanlikilicer;Ali Keles;A. Sima Uyar

  • Affiliations:
  • Istanbul Technical University, Istanbul, Turkey;Istanbul Technical University, Istanbul, Turkey;Istanbul Technical University, Istanbul, Turkey

  • Venue:
  • Proceedings of the 9th annual conference companion on Genetic and evolutionary computation
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Many real-world optimization problems are dynamic in nature. The interest in the Evolutionary Algorithms (EAs) community in applying EA variants to dynamic optimization problems has increased greatly. Differential Evolution (DE) belongs to the group of evolutionary algorithms which operate in continuous search spaces. DE has been successfully applied to many stationary problem domains. Recently there has been some research into applying DE to dynamic optimization problems too. Many real-world problems consist of decision variables which require the optimization algorithm to work with binary parameters. This makes it impossible to apply DE in its basic form. For this purpose, binary differential evolution (BDE) approaches have been introduced. The main focus of this paper is to perform a series of experiments to test the behavior of a simple BDE under different change conditions. A simple bit-matching problem is chosen as the test environment. The results of this preliminary study show that further study is needed to make BDEs suitable to work in dynamic environments.